JP2022181905A - Mechanical constant estimation device and motor control device - Google Patents

Abstract

To provide a mechanical constant estimation device and a motor control device that make it possible to estimate inertia, a coefficient of viscosity and Coulomb friction in a short time with high precision even when a motor speed and acceleration are improper for estimation.SOLUTION: A mechanical constant estimation device comprises: integration means 304 which performs numerical integration of products of two out of torque, acceleration, a speed, etc., of a motor in a period in which a speed absolute value is equal to or larger than a predetermined speed; estimation value computation means 311 which estimates inertia, a coefficient of viscosity and Coulomb friction based upon the integral value; integration means 306, 308 and 310 which compute weights; inertia estimation value average means 340 which uses an inertia estimation value and a weight to update an inertial estimation average value J2; viscosity coefficient estimation value average means 360 which uses a viscosity coefficient estimation value and a weight to update a viscosity estimation average value D2; and Coulomb friction estimation value average means 380 which uses a Coulomb friction estimation value and a weight to update a Coulomb estimation average value Tf2.SELECTED DRAWING: Figure 2

Description

The present invention relates to a machine constant estimating device for estimating mechanical constants of a motor and its mechanical load, and a motor control device equipped with this machine constant estimating device.

In order to control the speed of a motor that drives a mechanical load, the inertia of the motor and its mechanical load, the viscosity coefficient (viscous friction coefficient), the load torque such as Coulomb friction, etc. must be obtained and reflected in the speed control conditions. is required.
Also, even if the motor continues to drive the same mechanical load, the load torque may change. In such a case, it would be convenient to be able to estimate the mechanical constants such as inertia regardless of the operating conditions.

Here, in Non-Patent Document 1, on the premise that the load torque is represented by the sum of the speed-independent sign function type Coulomb friction and the viscosity coefficient proportional to the speed, the Coulomb friction, the viscosity coefficient, and the inertia A method for estimating is disclosed.
In this document, a positive/negative symmetric periodic signal is given as a speed command, and as shown in FIG. 8 (FIG. 4 of Non-Patent Document 1), signals τ _e , q ₀ , q ₀ ^′ , The values obtained by multiplying q ₁ with each other are integrated at periodic intervals according to the formulas (Eqs. (37) to (44)) described in the same document, and the elements φ ₁₁ , φ ₁₃ , φ ₂₂ , φ ₂₃ , φ ₃₃ and the elements v ₁ , v ₂ and v ₃ of the vector V are obtained, and then the calculation Φ ⁻¹ V is performed to identify the machine constants including inertia.

Further, Patent Document 1 describes a motor control device that identifies inertia and viscosity coefficients by performing adaptive identification calculations on torque command differential values, motor acceleration, motor jerk, etc. through low-pass filters having the same characteristics. Furthermore, in Patent Document 2, the velocity, acceleration, and torque weighted using a weight signal that is updated according to the time after the drive machine is started are input to a least-squares calculation unit to estimate inertia and viscosity coefficients. A machine constant identification device for a drive machine is described.

Japanese Patent Application Laid-Open No. 2006-217729 ([0036] to [0059], FIG. 1, etc.) Japanese Patent No. 3683121 ([0092] to [0103], FIG. 7, FIG. 8, etc.)

Ichiro Awaya et al., "Method for identifying parameters of a mechanical motion system in which Coulomb friction acts", Transactions of the Japan Society of Mechanical Engineers (Edition C), Vol.59, No.567 (1993), pp.108-114

According to the method of identifying the mechanical constants described in Non-Patent Document 1, for example, even when the motor is driven with a small acceleration that makes it difficult to obtain the desired inertia estimation accuracy, the inertia estimated value is updated regardless of the estimation accuracy. . Alternatively, even when the motor is driven at a small maximum speed at which it is difficult to obtain the desired estimation accuracy of the viscosity coefficient, the estimated inertia value is updated regardless of the estimation accuracy.
That is, even if a more reliable estimation result was obtained in the past, the estimated value may be updated to the latest less reliable estimation result. Furthermore, while it is desirable to be able to estimate the machine constants as accurately as possible, there are cases where it is desired to estimate the machine constants in a short period of time. It was difficult to reconcile the

Further, in the prior art disclosed in Patent Document 1, the inertia and the viscosity coefficient are identified at high speed and with high accuracy even when the rotational speed of the motor is low. During the period until elapses, the estimation error due to Coulomb friction is reduced by reducing the weight.
However, these Patent Documents 1 and 2 do not particularly disclose means for estimating Coulomb friction.

Therefore, the problem to be solved by the present invention is to be able to estimate the inertia, viscosity coefficient, and Coulomb friction with high accuracy and in a short time in a manner that does not cause deterioration of the estimated values even when the motor speed and acceleration are not appropriate for estimating the mechanical constants. and a motor control device for controlling the speed of a motor, etc., using the machine constants estimated by the estimating device.

In order to solve the above problems, a mechanical constant estimating device according to claim 1 is a mechanical constant estimating device that estimates mechanical constants consisting of inertia, viscosity coefficient, and Coulomb friction of a motor and a mechanical load driven by the motor,
After the motor is started, when the absolute value of the speed exceeds a predetermined estimated start speed, numerical integration is started using the product of any two of the torque, acceleration, speed, and speed sign of the motor as an integrand function. and integration means for terminating numerical integration when the motor decelerates and the absolute value of the speed becomes equal to or less than the estimated start speed;
estimated value calculation means for calculating an inertia estimated value J ₁ , a viscosity coefficient estimated value D ₁ , and a Coulomb friction estimated value T _f1 in the acceleration/deceleration period of the motor based on the integrated values obtained by the integration means;
means for calculating weights W _J1 , W _D1 , and W _Tf1 respectively corresponding to the estimated inertia value _J 1 , the estimated viscosity coefficient value D ₁ , and the estimated Coulomb friction value T _f1 ;
Using the estimated inertia value _J1 and the weight _WJ1 in the acceleration/deceleration period, update the estimated inertia average value _J2 and the integrated weight _WJ2 calculated before the acceleration/deceleration period, and update the estimated inertia average value J after the update. inertia estimation value averaging means for outputting ₂ as an inertia estimation result;
Using the viscosity coefficient estimated value _D1 and the weight W _D1 in the acceleration/deceleration period, the viscosity coefficient estimated average value _D2 and the integrated weight W _D2 calculated before the acceleration/deceleration period are updated, and the viscosity coefficient estimation after updating viscosity coefficient estimated value averaging means for outputting the average value _D2 as a viscosity coefficient estimation result;
Using the Coulomb friction estimated value T _f1 and the weight W _Tf1 in the acceleration/deceleration period, the Coulomb friction estimated average value T _f2 and the integrated weight W _Tf2 calculated before the acceleration/deceleration period are updated, and the updated Coulomb friction estimation Coulomb friction estimated value averaging means for outputting an average value T _f2 as a Coulomb friction estimation result;
characterized by comprising

A machine constant estimating device according to claim 2 is the machine constant estimating device according to claim 1,
The estimated value calculation means is
Acceleration a(t), velocity v(t), velocity sign sign(v(t)), and torque T(t) in the acceleration/deceleration period t of the motor are represented by x ₁ (t), x 2 (t), and x ₂ (t), respectively. When x ₃ (t) and y(t) are set, the product of any two of the above x ₁ (t), x ₂ (t), x ₃ (t) and y(t) is the integrand function Using m _ij = ∫ {x _i (t) x _j (t)} dt and q _i = ∫ {x _i (t) y (t)} dt (where i, j = 1 to 3), the patent The inertia estimated value J ₁ , the viscosity coefficient estimated value D ₁ , and the Coulomb friction estimated value T _f1 are calculated by Equation 1 described in the claims.

A machine constant estimating device according to claim 3 is the machine constant estimating device according to claim 1 or 2, wherein the torque, acceleration, and speed for obtaining the integrand are calculated by low-pass filters with the same time constant. It is characterized by being the latter value.

A mechanical constant estimating apparatus according to claim 4 is the machine constant estimating apparatus according to claim 1 or 2, wherein the weight _WJ1 is a value obtained by numerically integrating a value monotonically increasing with respect to the absolute value of the acceleration of the motor. and the weight W _D1 is a value obtained by numerically integrating a monotonically increasing value with respect to the absolute value of the speed of the motor, and the weight W _Tf1 does not depend on the speed or acceleration of the motor. It is characterized by being a value obtained by numerically integrating a value.

A machine constant estimating apparatus according to claim 5 is the machine constant estimating apparatus according to claim 3, wherein the weight _WJ1 is a value that monotonically increases with respect to the absolute value of the acceleration of the motor after calculation by the low-pass filter. is a value obtained by numerically integrating the weight W D1 , the weight W _D1 is a value obtained by numerically integrating a value monotonically increasing with respect to the absolute value of the speed of the motor after calculation by the low-pass filter, and the weight W _Tf1 is , is a value obtained by numerically integrating a value that does not depend on the speed or acceleration of the motor.

A mechanical constant estimating device according to claim 6 is the mechanical constant estimating device according to claim 4 or ₅ , wherein the inertia estimated value averaging means is configured _to a value obtained by limiting the sum of the previous value of the integrated weight _WJ2 and the current value of the weight _WJ1 by a predetermined upper limit value _WJmax as the current value of the integrated weight _WJ2 ;
Current value of estimated inertia average value _J2 = Previous value of estimated inertia average value _J2 + (current value of weight W _J1 / current value of cumulative weight W _J2 ) × (current value of estimated inertia value _J1 - estimated inertia average value previous value of value _J2 )
is characterized by computing

A mechanical constant estimating device according to claim 7 is the mechanical constant estimating device according to any one of claims 4 to 6, wherein said viscosity coefficient estimated value averaging means comprises said viscosity coefficient estimated value _D1 and said weight W Each time _D1 is newly obtained, a value obtained by limiting the sum of the previous value of the cumulative weight _WD2 and the current value of the weight _WD1 by a predetermined upper limit value _WDmax is used as the current value of the cumulative weight _WD2 . After that,
Current value of estimated average value of viscosity coefficient _D2 = Previous value of estimated average value of viscosity coefficient _D2 + (Current value of weight W _D1 / Current value of integrated weight W _D2 ) × (Current value of estimated viscosity coefficient _D1 - Previous value of viscosity coefficient estimated average value _D2 )
is characterized by computing

A mechanical constant estimating device according to claim 8 is the mechanical constant estimating device according to any one of claims 4 to 7, wherein said Coulomb friction estimated value averaging means comprises said Coulomb friction estimated value T _f1 and said weight W Each time _Tf1 is newly obtained, a value obtained by limiting the sum of the previous value of the cumulative weight W _Tf2 and the current value of the weight W _Tf1 by a predetermined upper limit value W _Tfmax is used as the current value of the cumulative weight W _Tf2 . After that,
Current value of Coulomb friction estimated average value T _f2 = Previous value of Coulomb friction estimated average value T _f2 + (Current value of weight W _Tf1 / Current value of integrated weight W _Tf2 ) × (Current value of Coulomb friction estimated value T _f1 − Previous value of Coulomb friction estimated average value T _f2 )
is characterized by computing

A mechanical constant estimating device according to claim 9 is the mechanical constant estimating device according to any one of claims 1 to 8, wherein the inertia estimated average value J ₂ , the viscosity coefficient estimated average value D ₂ and the Coulomb friction The estimated average value T _f2 and the integrated weights W _J2 , W _D2 , and W _Tf2 are stored in a non-volatile memory and can be initialized from the outside.

A motor control device according to claim 10 is characterized in that the motor is controlled using the machine constant estimated by the machine constant estimation device according to any one of claims 1 to 9.

According to the present invention, even if it is difficult to properly estimate the machine constants because the motor speed or acceleration is small, it is possible to estimate the machine constants with high accuracy and in a short period of time in such a way that the estimated values are less likely to deteriorate. .

1 is a block diagram showing an overview of a machine constant estimating device according to an embodiment of the present invention; FIG. FIG. 2 is a configuration diagram showing a first embodiment of a machine constant estimator in FIG. 1; 4 is a configuration diagram of integration start/end determination means in each embodiment of the machine constant estimator; FIG. FIG. 2 is a configuration diagram showing a second embodiment of a machine constant estimator in FIG. 1; 4 is a configuration diagram of inertia estimated value averaging means in each embodiment of the machine constant estimator; FIG. FIG. 4 is a block diagram of viscosity coefficient estimated value averaging means in each embodiment of the mechanical constant estimator; FIG. 4 is a configuration diagram of Coulomb friction estimated value averaging means in each embodiment of the machine constant estimator; FIG. 3 is an explanatory diagram of the conventional technology described in Non-Patent Document 1;

An embodiment of the present invention will be described below with reference to the drawings.
FIG. 1 is a block diagram showing an outline of a machine constant estimation device according to this embodiment. This mechanical constant estimator estimates the inertia, viscosity coefficient, and Coulomb friction of the motor/mechanical load 200 based on the speed and torque of the motor to which the mechanical load is connected (the two are collectively referred to as the motor/mechanical load 200). do.

In FIG. 1, the speed control unit 100 outputs torque so that the motor speed obtained from the motor/mechanical load 200 matches the speed command. Here, the torque may be a torque command, or may be a torque estimated value obtained by detecting the current supplied to the motor.
Motor speed and torque are input to the machine constant estimator 300 . The mechanical constant estimator 300 estimates the inertia, the viscosity coefficient, and the Coulomb friction by operations described later, and performs weighting processing to calculate these average values J ₂ , D ₂ , and T _f2 .
Note that the speed control unit 100 and the machine constant estimation unit 300 are implemented by an arithmetic processing unit such as a microcomputer and its program, for example.

FIG. 2 is a configuration diagram showing a first embodiment of the machine constant estimator 300 (machine constant estimator 300A).
In FIG. 2, the motor speed obtained from the motor/mechanical load 200 is given to the integration start/end determination means 320, the differentiation means 301, and the sign determination means 302 in the machine constant estimator 300A, and the motor torque is multiplied. provided to means 303; A reset command from the outside is input to the inertia estimated value averaging means 340, the viscosity coefficient estimated value averaging means 360, and the Coulomb friction estimated value averaging means 380, which will be described later.

Let a(t) be the motor acceleration, v(t) be the motor speed, sign(v(t)) be the sign of the motor speed, and T ₍ t) be the torque _. (t), x ₃ (t), and y(t), motor acceleration x ₁ (t) which is the output of differentiation means 301, motor speed x ₂ (t), and velocity which is the output of sign determination means 302 Sign x ₃ (t) and torque y(t) are input to multiplier 303 .

Multiplication means 303 uses two parameters out of x ₁ (t), x ₂ (t), x ₃ (t) and y(t) to obtain x _i (t) x _j (t) and x _i ( t) Calculate y(t) (i, j=1-3). Note that calculations in the multiplication means 303 include calculations of x ₁ (t) ² , x ₂ (t) ² , and x ₃ (t) ² (that is, i=j=1, i=j=2, i=j =3).
The results of multiplication by the multiplication means 303 are input to the integration means 304 in the next stage.

Integrating means 304 performs numerical integration of m _ij =∫{x _i (t)x _j (t)}dt and q _i =∫{x _i (t)y(t)}dt over a certain period of time, These calculation results are sent to the estimated value calculation means 311 .
In actual numerical integration, for example,
m _ij =Σ{x _i (n)x _j (n)},q _i =Σ{x _i (n)y(n)}
Define m _ij and q _i by performing the operation, and when the n-th data is obtained,
m _ij ←m _ij +x _i (n)x _j (n),
q _i ← q _i + x _i (n)y(n)
, update m _ij and q _i respectively.

FIG. 3 is a block diagram of the integration start/end determination means 320 in FIG.
In FIG. 3, the absolute value computing means 321 computes the absolute value of the motor speed and sends it to the comparing means 322, and the comparing means 322 compares the preset estimated start speed and the speed absolute value.
Then, when the speed absolute value exceeds the estimated start speed, an integration start instruction is sent from the comparison means 322 to the integration means 304 in FIG. At this time, the comparison means 322 sends an integration end instruction to the integration means 304 via the instruction generation means 324 . The period from the integration start instruction to the integration end instruction is the integration period. Further, from the _instruction generating means ₃₂₄ , the estimated value calculation means 311 of _FIG . An instruction to update the weights W _J1 , W _D1 , and W _Tf1 for the viscosity coefficient estimated value averaging means 360 and the Coulomb friction estimated value averaging means 380 is also output.

The reasons why the timings of the integration start instruction/integration end instruction and update instruction by the integration start/end determination means 320 are set as described above are as follows.
When a motor drives a mechanical load, it often produces load torque with hysteresis near zero speed. In order for the estimated value calculation means 311 to correctly estimate the machine constant, it is necessary to exclude the above hysteresis region from the period during which the integration means 304 performs integration. Therefore, integration is started after the absolute speed value exceeds a predetermined estimated start speed after the motor starts, and the integration is stopped when the motor decelerates and the absolute speed value becomes equal to or less than the estimated start speed. and instructing to update each parameter, the inertia estimated value J ₁ , the viscosity coefficient estimated value D ₁ , the Coulomb friction estimated value T _f1 , and the weights W J1 , W D1 , W _J1 , W _D1 , It was decided to determine _WTf1 .
In this case, by equalizing the motor speed absolute values at the integration end instruction time and the integration start time point (that is, using the same estimated start speed), the error in each estimated value when there is nonlinearity in the load torque is reduced. can do.

If it is problematic to use the speed signal, its differential acceleration signal, and torque signal as they are due to detection errors in the motor speed and current, the second embodiment shown in FIG. As in the machine constant estimating section 300B in the example, it is preferable to use the results obtained by calculating the motor speed and torque with the low-pass filters 312 and 313, respectively. In this case, it is desirable that the time constants of the low-pass filters 312 and 313 be equal.

Returning to FIG. 2, the estimated value calculation means 311 uses the m _ij and q _i (i, j=1 to 3) calculated by the integration means 304 to calculate the inertia estimated value J ₁ , the viscosity coefficient estimated value D ₁ , and the Coulomb friction estimated value T _f1 are calculated by Equation (1).

Equation 1 is a system whose equation of motion is expressed as T(t)=J(dv/dt)+Dv+T _f sign(v), and parameters J ₁ , D ₁ , This corresponds to obtaining T _f1 .

The above will be specifically explained. In the least squares method,
Integral value I=∫{T(t)-J(dv/dt)-Dv-T _f sign(v)} ² dt
Find J, D, and T _f that minimize For this reason, if the values obtained by partially differentiating the integrated value I with respect to J, D, and _Tf are set to zero, Equation 2 is obtained.

数式３における積分演算∫ｄｔを積算演算Σに置き換えると、数式４が得られる。

Equation 3 is obtained by replacing Equation 2 above with x ₁ (t), x ₂ (t), x ₃ (t), and y(t).

Replacing the integration operation ∫dt in Equation 3 with the integration operation Σ yields Equation 4.

この数式５を変形すると数式６が得られ、数式６におけるＪをＪ_１、ＤをＤ_１、Ｔ_ｆをＴ_ｆ１とおけば、前述の数式１によって慣性推定値Ｊ_１、粘性係数推定値Ｄ_１、及びクーロン摩擦推定値Ｔ_ｆ１を求めることができる。

Furthermore _, the n-th data _{x i ( n ) ,} x _j Replacing (n) and y(n) with data x _i (t), x _j (t) and y(t) at time t respectively, mi _ij =Σ{x _i (t)x _j (t)} , q _i =Σ{x _i (t)y(t)}, so Equation 4 can be expressed as Equation 5.

Equation 6 is obtained by modifying Equation 5. If J is J ₁ , D is D ₁ , and T _f is T _f1 in Equation 6, the estimated inertia value J ₁ and the estimated viscosity coefficient D are obtained from Equation 1 above. ₁ and the Coulomb friction estimate T _f1 can be determined.

On the other hand, as shown in FIG. 2, the motor acceleration x ₁ (t) is input to the first integrand value generating means 305 to generate the integrand value f ₁ that monotonously increases with respect to the absolute value of x ₁ (t). The motor speed x ₂ (t) is input to the second integrand generating means 307 to generate an integrand f ₂ that monotonically increases with respect to the absolute value of x ₂ (t). The third integrand generating means 309 generates an integrand _f3 that does not depend on the acceleration or speed of the motor. Here, for example, f ₁ may be the square of x ₁ (t), f ₂ may be the square of x ₂ (t), and f ₃ may simply be 1.

The integrands f ₁ , f ₂ , and f ₃ are integrated by integrating means 306 , 308 , and 310 , respectively, and the results are weighted as weights W _J1 , W _D1 , and W _Tf1 . It is input to the averaging means 360 and the Coulomb friction estimated value averaging means 380, respectively.
The inertia estimated value averaging means 340, the viscosity coefficient estimated value averaging means 360, and the Coulomb friction estimated value averaging means 380 receive the inertia estimated value J ₁ , the viscosity coefficient estimated value D ₁ , and the Coulomb friction Estimated average values J 2 , D 2 , and T _f2 are calculated by performing weighting operations using weights W _{J1 ,} W _D1 , _and W _Tf1 on the estimated value T _f1 as follows.

The configurations and operations of the inertia estimated value averaging means 340, the viscosity coefficient estimated value averaging means 360, and the Coulomb friction estimated value averaging means 380 will be described below.
For example, when the inertia is small and the acceleration is also small, even if the estimation calculation by the estimated value calculation means 311 is performed, it is conceivable that the inertia cannot be estimated with sufficient accuracy with only one acceleration/deceleration operation. Also, even when acceleration/deceleration operation with a small maximum speed is performed, it is considered that the viscosity coefficient cannot be estimated with sufficient accuracy with only one acceleration/deceleration operation. It is conceivable that the Coulomb friction cannot be estimated with sufficient accuracy only by the single acceleration/deceleration operation.

When the acceleration/deceleration operation of a similar pattern is repeated, the estimation accuracy can be improved by applying a simple moving average to each estimated value obtained for each acceleration/deceleration operation. However, if it is desired to improve the accuracy of estimating each machine constant in a non-repetitive operation, the reliability of each estimation will differ for each acceleration/deceleration operation.
Therefore, in this embodiment, the following operations by the inertia estimated value averaging means 340, the viscosity coefficient estimated value averaging means 360, and the Coulomb friction estimated value averaging means 380 improve the estimation accuracy.

First, the inertia estimated value averaging means 340 will be described with reference to FIG.
It is assumed that the switching means 344 and 348 are in the illustrated state until the reset command is input. This point also applies to switching means 364 and 368 in FIG. 6 and switching means 384 and 388 in FIG. 7, which will be described later.

In FIG. 5, the estimated inertia value averaging means 340 updates the average estimated inertia value J2 and the weight _WJ2 each _time the estimated inertia value _J1 and the weight _WJ1 are newly obtained in one acceleration/deceleration period. works. Note that the weight W _J2 is always obtained as a value obtained by adding the current value (latest value) and the previous value of the weight W _J1 , so hereinafter, it is referred to as "integrated weight W _J2 ".
As shown in FIG. 5, the previous value of integrated weight _WJ2 held by previous value holding means 343 and the current value (latest value) of weight _WJ1 are added by addition/subtraction means 341, and the result is The current value of the integrated weight _WJ2 is obtained by limiting it with the upper limit value _WJmax . Then, the division means 342 divides the current value of the weight _WJ1 by the current value of the integrated weight _WJ2 , and the result is input to the multiplication means 346 .

Further, the deviation between the previous value of the estimated inertia average value _J2 held by the previous value holding means 349 and the current value of the estimated inertia value _J1 is obtained by the addition/subtraction means 345, and the deviation is input to the multiplication means 346 for division. Multiply by the output of means 342 . Furthermore, the output of the multiplication means 346 and the previous value of the estimated inertia average value _J2 are added by the addition/subtraction means 347, and the result is output as the current value of the estimated inertia average value _J2 .

The above operation is expressed by a formula as follows.
_J2 current value = _J2 previous value + (W _J1 current value / W _J2 current value) x ( _J1 current value - _J2 previous value)
Note that the integrated weight W _J2 and the inertia estimated average value _J2 are substantially controlled by switching the switching means 344 and 348 to the "0" side in response to a reset command from the outside. can be initialized by removing Specifically, for example, when the mechanical load is changed, the switching means 344 and 348 are switched to the "0" side to initialize W _J2 and J ₂ , and after that W _J2 and J ₂ are stored in storage means such as a non-volatile memory. Update while remembering. Also, when the system is restarted, the values of W _J2 and J ₂ are read from the storage means prior to operation after restart.

A method for updating the estimated inertia average value _J2 will be described in more detail.
First, when the value obtained by adding the previous value of _WJ2 and the current value of _WJ1 by the addition/subtraction means 341 remains below the upper limit value _WJmax ,
_J2 current value = _J2 previous value + {W _J1 current value / (W _J2 previous value + W _J1 current value)} x ( _J1 current value - _J2 previous value)
= ( _J2 previous value x W _J2 previous value + _J1 current value x W _J1 current value) / (W _J2 previous value + W _J1 current value)
becomes.
This is equivalent to obtaining a weighted average value using all the data of past acceleration/deceleration. Since x ₁ (t) in FIG. 2 is reduced and the weight W _J1 is kept small, a highly reliable estimated inertia average value J ₂ can be obtained.

However, if the operation is continued without any restriction on the value obtained by adding the previous value of _WJ2 and the current value of _WJ1 , _WJ2 eventually becomes infinite and the output of the dividing means 342 approaches zero. Therefore, even if a new acceleration/deceleration operation is performed, the estimated inertia average value _J2 will not be updated.
Therefore, as shown in FIG. 5, by limiting the result of addition of the previous value of _WJ2 and the current value of _WJ1 to an upper limit value _WJmax or less, the latest operation result will increase as the acceleration/deceleration operation continues. It was made to be reflected in the inertia estimated average value _J2 .

Now, without defining weights,
_J2 current value = _J2 previous value + constant x ( _J1 current value - _J2 previous value) (0 < constant < 1)
Consider the case where the _J2 current value is obtained by
Even when calculating in this way, the accuracy of inertia estimation increases as the acceleration/deceleration operation is repeated. However, in this case, the current _J2 value becomes smaller than the true value until the number of acceleration/deceleration operations reaches about (3/constant). In addition, even when low acceleration or short-time acceleration/deceleration operation that is not suitable for inertia estimation is performed, the estimated inertia value will be updated based on the result in the same manner as other driving patterns.

On the other hand, in the present embodiment, the values of W _J2 and J ₂ are initialized by a reset command when the mechanical load is replaced, for example, so that even if the number of acceleration/deceleration operations is small, the estimated inertia value is close to the true value. is obtained, and a highly accurate estimated inertia average value _J2 can be obtained in a short time.
Further, the weight W _J1 is obtained by numerically integrating f ₁ monotonously increasing with respect to the absolute value of the acceleration x ₁ (t), and the inertia estimated value J ₁ is updated using this weight W _J1 and the integrated weight W _J2 . Therefore, the estimated inertia average value _J2 is less likely to be disturbed even for driving that is not suitable for inertia estimation.

Next, FIG. 6 shows the configuration of the viscosity coefficient estimated value averaging means 360. As shown in FIG.
This viscosity coefficient estimated value averaging means 360 updates the viscosity coefficient estimated _{average value D2 and integrated weight W D2 each time a new viscosity coefficient estimated value D1 and weight W D1 are obtained in} one acceleration/deceleration period. works like

In the viscosity coefficient estimated value averaging means 360 shown in FIG. 6, 361, 365 and 367 are addition/subtraction means, 362 is division means, 363 and 369 are previous value holding means, 364 and 368 are switching means, 366 is multiplication means, and _WDmax . is the upper limit set to the addition result of the previous value of _WD2 and the current value of _WD1 . The overall operation of this viscosity coefficient estimate averaging means 360 is similar to the inertia estimate averaging means 340 except for input and output signals.
In estimating the viscosity coefficient, the viscosity coefficient estimated value D1 is calculated using the weight W _D1 obtained by numerically integrating _f2 , which monotonically increases with respect to the absolute value of the speed _x2 (t), and the integrated weight W _{D2 .} Since the estimated viscosity coefficient average value _D2 is obtained by updating, the estimated viscosity coefficient average value _D2 is less likely to be disturbed even for driving patterns that are not suitable for estimating the viscosity coefficient, such as driving with a low maximum speed. .

7 shows the configuration of the Coulomb friction estimated value averaging means 380. As shown in FIG.
This Coulomb friction estimated value averaging means 380 updates the Coulomb friction estimated average value T f2 and _{the integrated weight W Tf2 each time the Coulomb friction estimated value T f1 and the weight W Tf1 are newly obtained} in one acceleration/deceleration period. works like

In the Coulomb friction estimated value averaging means 380 shown in FIG. 7, 381, 385 and 387 are addition/subtraction means, 382 is division means, 383 and 389 are previous value holding means, 384 and 388 are switching means, 386 is multiplication means, and W _Tfmax . is the upper limit set to the addition result of the previous value of W _Tf2 and the current value of W _Tf1 . The overall operation of this Coulomb friction estimated value averaging means 380 is similar to the inertia estimated value averaging means 340 and the viscosity coefficient estimated value averaging means 360 described above, except for the input/output signals.
In estimating the Coulomb friction, the weight W _Tf1 obtained by numerically integrating the value _f3 that does not depend on either the acceleration x ₁ (t) or the velocity x ₂ (t) and the integrated weight W _Tf2 are used to estimate the Coulomb friction. Since the Coulomb friction estimated average value T _f2 is obtained by updating the value T _f1 , the Coulomb friction estimated average value T _f2 is less likely to be disturbed even when the operation is not suitable for Coulomb friction estimation.

As described above, according to the machine constant estimating apparatus of the present embodiment, even if the motor speed, acceleration, and acceleration/deceleration operation period are inappropriate for estimating the machine constant, the estimated value deteriorates. It is possible to estimate the machine constants with high accuracy and in a short time in a form that is less likely to cause
Furthermore, by implementing this machine constant estimating device in the motor control device and reflecting the machine constant estimated for each acceleration/deceleration period according to the mechanical load in the speed control and torque control of the motor, it is possible to reduce the following error in the mechanical drive system. Vibration can be reduced.

100: Speed control unit 200: Motor/mechanical load 300, 300A, 300B: Machine constant estimation unit 301: Differentiation means 302: Sign determination means 303: Multiplication means 304, 306, 308, 310: Integration means 305, 307, 309: Integral value generating means 311: Estimated value computing means 312, 313: Low-pass filter 320: Integration start/end determining means 321: Absolute value computing means 322: Comparing means 323, 324: Instruction generating means 340: Inertia estimated value averaging means , 345, 347: addition/subtraction means 342: division means 343, 349: previous value holding means 346: multiplication means 344, 348: switching means 360: viscosity coefficient estimated value average means 361, 365, 367: addition/subtraction means 362: division means , 369: previous value holding means 366: multiplication means 364, 368: switching means 380: Coulomb friction estimated value averaging means 381, 385, 387: addition/subtraction means 382: division means 383, 389: previous value holding means 386: multiplication means 384 , 388: switching means

Claims (10)

In a mechanical constant estimating device for estimating mechanical constants consisting of inertia, viscosity coefficient, and Coulomb friction of a motor and a mechanical load driven by the motor,
After the motor is started, when the absolute value of the speed exceeds a predetermined estimated start speed, numerical integration is started using the product of any two of the torque, acceleration, speed, and speed sign of the motor as an integrand function. and integration means for terminating numerical integration when the motor decelerates and the absolute value of the speed becomes equal to or less than the estimated start speed;
estimated value calculation means for calculating an inertia estimated value J ₁ , a viscosity coefficient estimated value D ₁ , and a Coulomb friction estimated value T _f1 in the acceleration/deceleration period of the motor based on the integrated values obtained by the integration means;
means for calculating weights W _J1 , W _D1 , and W _Tf1 respectively corresponding to the estimated inertia value _J 1 , the estimated viscosity coefficient value D ₁ , and the estimated Coulomb friction value T _f1 ;
Using the estimated inertia value _J1 and the weight _WJ1 in the acceleration/deceleration period, update the estimated inertia average value _J2 and the integrated weight _WJ2 calculated before the acceleration/deceleration period, and update the estimated inertia average value J after the update. inertia estimation value averaging means for outputting ₂ as an inertia estimation result;
Using the viscosity coefficient estimated value _D1 and the weight W _D1 in the acceleration/deceleration period, the viscosity coefficient estimated average value _D2 and the integrated weight W _D2 calculated before the acceleration/deceleration period are updated, and the viscosity coefficient estimation after updating viscosity coefficient estimated value averaging means for outputting the average value _D2 as a viscosity coefficient estimation result;
Using the Coulomb friction estimated value T _f1 and the weight W _Tf1 in the acceleration/deceleration period, the Coulomb friction estimated average value T _f2 and the integrated weight W _Tf2 calculated before the acceleration/deceleration period are updated, and the updated Coulomb friction estimation Coulomb friction estimated value averaging means for outputting an average value T _f2 as a Coulomb friction estimation result;
A machine constant estimating device comprising: In the machine constant estimation device according to claim 1,
The estimated value calculation means is
Acceleration a(t), velocity v(t), velocity sign sign(v(t)), and torque T(t) in the acceleration/deceleration period t of the motor are represented by x ₁ (t), x 2 (t), and x ₂ (t), respectively. When x ₃ (t) and y(t) are set, the product of any two of the above x ₁ (t), x ₂ (t), x ₃ (t) and y(t) is the integrand function Using m _ij = ∫ {x _i (t) x _j (t)} dt and q _i = ∫ {x _i (t) y (t)} dt (where i, j = 1 to 3), the following A machine constant estimating device, which calculates an inertia estimated value J ₁ , a viscosity coefficient estimated value D ₁ , and a Coulomb friction estimated value T _f1 according to Equation 1 of .

In the machine constant estimation device according to claim 1 or 2,
A machine constant estimating device, wherein the torque, acceleration, and speed for obtaining the integrand are values obtained after calculating a low-pass filter with the same time constant. In the machine constant estimation device according to claim 1 or 2,
The weight W _J1 is a value obtained by numerically integrating a value that monotonously increases with respect to the absolute value of the acceleration of the motor, and the weight W _D1 is a value that is numerically integrated with a value that monotonously increases with respect to the absolute value of the speed of the motor. A machine constant estimating device, wherein the weight W _Tf1 is a value obtained by numerically integrating a value independent of the speed and acceleration of the motor. In the machine constant estimation device according to claim 3,
The weight W _J1 is a value obtained by numerically integrating a monotonically increasing value with respect to the absolute value of the acceleration of the motor after calculation by the low-pass filter, and the weight W _D1 is the motor after calculation by the low-pass filter. and the weight W _Tf1 is a value obtained by numerically integrating a value that does not depend on the speed or acceleration of the motor. , the machine constant estimator. In the machine constant estimation device according to claim 4 or 5,
The inertia estimate averaging means comprises:
Each time the estimated inertia value _J1 and the weight _WJ1 are newly obtained, the sum of the previous value of the integrated weight _WJ2 and the current value of the weight _WJ1 is limited by a predetermined upper limit value _WJmax . is the current value of the integrated weight _WJ2 ,
Current value of estimated inertia average value _J2 = Previous value of estimated inertia average value _J2 + (current value of weight W _J1 / current value of cumulative weight W _J2 ) × (current value of estimated inertia value _J1 - estimated inertia average value previous value of value _J2 )
A machine constant estimating device characterized by computing In the machine constant estimation device according to any one of claims 4 to 6,
The viscosity coefficient estimated value averaging means
Each time the viscosity coefficient estimated value _D1 and the weight _WD1 are newly obtained, the sum of the previous value of the integrated weight _WD2 and the current value of the weight _WD1 is limited by a predetermined upper limit value _WDmax . After setting the value as the current value of the integrated weight W _D2 ,
Current value of estimated average value of viscosity coefficient _D2 = Previous value of estimated average value of viscosity coefficient _D2 + (Current value of weight W _D1 / Current value of integrated weight W _D2 ) × (Current value of estimated viscosity coefficient _D1 - Previous value of viscosity coefficient estimated average value _D2 )
A machine constant estimating device characterized by computing In the machine constant estimation device according to any one of claims 4 to 7,
The Coulomb friction estimate averaging means,
Each time the Coulomb friction estimated value T _f1 and the weight W _Tf1 are newly obtained, the sum of the previous value of the integrated weight W _Tf2 and the current value of the weight W _Tf1 is limited by a predetermined upper limit value W _Tfmax . After setting the value to the current value of the integrated weight W _Tf2 ,
Current value of Coulomb friction estimated average value T _f2 = Previous value of Coulomb friction estimated average value T _f2 + (Current value of weight W _Tf1 / Current value of integrated weight W _Tf2 ) × (Current value of Coulomb friction estimated value T _f1 − Previous value of Coulomb friction estimated average value T _f2 )
A machine constant estimating device characterized by computing In the machine constant estimation device according to any one of claims 1 to 8,
The inertia estimated average value J ₂ , the viscosity coefficient estimated average value D ₂ , the Coulomb friction estimated average value T _f2 , and the integrated weights W _J2 , W _D2 , and W _Tf2 are stored in a nonvolatile memory and externally A machine constant estimating device characterized by being able to be initialized. A motor control device that controls the motor using the machine constant estimated by the machine constant estimation device according to any one of claims 1 to 9.

Images